Of the roughly 900,000 Americans who are currently infected with HIV, as many as 280,000 remain unaware of their infection and without access to life-prolonging services that could prevent the further spread of the virus. In 2002, we were awarded funding to develop a computer simulation model of HIV counseling, testing, and referral (HIVCTR) and the clinical management and economics of HIV. Over the last 2 years, we have achieved our project objectives. Our aims, in this continuation proposal, are: 1. To refine our simulation model of HIVCTR system performance and the impact of decisions on patient quality of life and behavior. 2. To conduct a series of policy evaluations on the value and design of expanded HIVCTR services. We will: 1. Define the circumstances under which newly approved, rapid HIV testing protocols are preferred to traditional antibody tests; 2. Estimate the long-run benefits and costs of real-time screening for primary HIV infection; 3. Examine how HIV testing outcomes translate into population-level changes in disease incidence and explore the possible impact of reduced transmission on estimates of the cost-effectiveness of expanded HIVCTR; 4. Project the downstream financial impact of increased case identification and referral to care on the demand for publicly funded HIV services, including Medicaid, the Ryan White CARE Act, and state AIDS Drug Assistance Programs (ADAPS). Over the last 10 years, our research team has established a record of publishing and disseminating findings that inform national priority setting in HIV patient care, clinical trial design, and practice guideline development. During the initial funding cycle of this project, we have demonstrated our potential to exert an equally profound impact in the area of HIV testing and prevention policy.